A forecaster is assessing two different models for demand. The output from each model and the actual demand data appear in the table. Use MAD and a tracking signal to compare the two models. Which model does a better job of forecasting?

Demand Model 1 Model 2
52 55.0 51.0
52 54.7 51.9
60 54.4 52.0
56 55.0 59.2
58 55.1 56.3
58 55.4 57.8
52 55.6 58.0
57 55.3 52.6
53 55.4 56.6
57 55.2 53.4
What will be an ideal response?

Answer: Based on calculations given in the table, it appears that Model 1 does a slightly better job of predicting Demand based on a lower MAD, 2.7 versus 3.2 for Model 2. Tracking signals for both models fall within the plus or minus 4 control limits suggested in the text.

Demand Model 1 Model 2 AD 1 AD 2 Tracking 1 Tracking 2 FE 1 FE 2
52 55.0 51.0 3.0 1.0 -1.0 1.0 -3.0 1.0
52 54.7 51.9 2.7 0.1 -2.0 2.0 -2.7 0.1
60 54.4 52.0 5.6 8.0 0.0 3.0 5.6 8.0
56 55.0 59.2 1.0 3.2 0.3 1.9 1.0 -3.2
58 55.1 56.3 2.9 1.7 1.2 2.7 2.9 1.7
58 55.4 57.8 2.6 0.2 2.2 3.3 2.6 0.2
52 55.6 58.0 3.6 6.0 0.9 0.6 -3.6 -6.0
57 55.3 52.6 1.7 4.4 1.6 2.0 1.7 4.4
53 55.4 56.6 2.4 3.6 0.7 0.8 -2.4 -3.6
57 55.2 53.4 1.8 3.6 1.4 2.0 1.8 3.6
MAD 2.7 3.2